Polypeptide preparation and characterization

Details can be found in [217].

3.4.2 Liposome preparation

Liposomes were prepared as described in section 2.4.5 except for the following modifications:

SUVs were prepared from DOPG, DOPS, DOPE, and DOPC lipids in aqueous 100 mM NaCl, 10 mM HEPES (pH 7.4).

3.4.3 SAXS experiments and data analysis

SAXS experiments and data analysis for polypeptide–membrane interactions were performed as described in section 2.4.7 except for the following modifications:

All polypeptide–lipid samples were prepared in 100 mM NaCl, 10 mM HEPES (pH 7.4). SAXS experiments were conducted at the Stanford Synchrotron Radiation Lightsource (SSRL, beamline 4–2) using monochromatic X-rays with an energy of 9 keV. The scattered radiation was collected using a Rayonix MX225-HE detector (pixel size of 73.2 μm).

3.5 Acknowledgments

We thank Prof. Paul J. Hergenrother for kindly supplying bacterial strains. J.C. acknowledges support from the NSF (Grant CHE-1153122) for the design and synthesis of polypeptide and the NIH (Director’s New Innovator Awards 1DP2OD007246 and 1R21EB013379) for the biological evaluation of the polypeptides. L.-F.C. and J.C. acknowledge support from NIH Grant R21AI117080 for antibacterial peptides against H. pylori. G.C.L.W. and M.W.L. acknowledge support from NSF Grant DMR-1411329 for the characterization of polypeptide–membrane interactions. R.M.P. acknowledges support from NIH Grants R01 DK 58587, R01 CA 77955, P01 CA 116087, and P30 DK 058404.

Chapter 4

Interactions between Membranes and “Metaphilic”

Polypeptide Architectures with Diverse Side-Chain

Populations

4.1 Introduction

The functions of proteins or peptides, such as molecular recognition, enzymatic reactions, and allosteric regulation, are determined by their structures and their internal motions: Proteins or peptides can fold into structures that present specific chemical patterns on their molecular surfaces. These nanoscopically defined patterns of charge, hydrogen bonding, and/or hydrophobicity, which strongly influence peptide or protein interactions, are known to be partly smeared out by thermal fluctuations. For protein configurations structurally cognate to the native folded state, the protein energy surface, which controls protein dynamics, can have multiple minima, and proteins exhibit harmonic motions within these minima as well as crossing of potential barriers between them. In general, however, molecular thermal motions are not large compared to the dimensions of the molecule. Moreover, single-molecule experiments show that folded proteins typically have Young’s moduli of ~1 × 108 _{Pa [219–221], which give the protein a solid-like rigidity. There is a}

rich literature showing that in the low-temperature limit, proteins can undergo a dynamic transition

This chapter is adapted with permission from Lee, M.W., Han, M., Bossa, G.V., Snell, C., Song, Z., Tang, H., Yin, L., Cheng, J., May, S., Luijten, E. & Wong, G.C.L. Interactions between Membranes and “Metaphilic” Polypeptide

to a glass-like solid state with small fluctuations [222, 223]. Taken together, the surface structure, shape, and elasticity of a protein determine the resultant presentation of surface chemistry, and thereby enable or limit its interactions. It would be interesting to start with the functional requirements for a given protein or peptide class, and explore the opposite limit, where patches of surface chemistry can be mobile.

Both AMPs and CPPs are short (generally < 50 amino acids) peptides that exert their functions by interacting with and permeating membranes. As part of the innate host defense, AMPs collectively exhibit broad-spectrum antimicrobial activity [12, 115, 117], typically through the disruption and permeabilization of bacterial membranes [115, 117, 224, 225]. Although AMPs are abundant and diverse in sequence and structure, they share some common features. Most AMPs are cationic and characterized by facially amphiphilic patterns of hydrophobicity and charge [12, 13, 115, 117, 224–226]. CPPs are capable of efficiently translocating across cell membranes and can mediate the uptake of conjugated cargos [227–229]. CPPs are generally cationic, but can also be amphiphilic, with the arginine-rich CPPs comprising the most widely studied group [227, 229– 233]. Many CPP sequences are derived from natural proteins and peptides; however, research groups have also developed synthetic CPPs [227, 232]. While cationic charge and amphiphilicity are characteristics often found in both AMPs and CPPs, it has been noted that these properties can be found in many other membrane-remodeling peptides [234], including viral budding peptides [43] and viral fusion peptides [235]. While the vast majority of AMPs and CPPs are composed of linear amino acid sequences, a number of research groups have recently explored unconventional nanoscopic architectures in the design of polymer-based antimicrobial and cell-penetrating agents that are also characterized by cationic charge and hydrophobicity, including circular peptides, and

236–240]. In this chapter, we systematically investigate a prototypical class of peptides with side- chain-rich architectures. These peptides consist of a rigid helical core decorated with mobile and flexible side chains that are terminated by cationic and hydrophobic groups, an arrangement that allows cationic and hydrophobic end groups to undergo large displacements, reminiscent of the Lindemann criterion for melting [241–243]. Therefore, these molecules have unusually chemically adaptable and quasi-liquid10 _{surfaces. Although one might expect that the loss of well-defined}

spatial relations between cationic and hydrophobic patches on a highly evolved peptide or protein leads to a degradation of activity, we surprisingly find the opposite. We show that the membrane- permeating activity of AMPs and CPPs, both commonly characterized by anchored cationic and hydrophobic groups, can be significantly enhanced by the highly adaptable side-chain-rich architecture: Like organisms that adapt to different colored environments via metachrosis, these molecular architectures adapt to different solvent environments (water, amphiphilic interface, hydrophobic membrane core) by being “metaphilic” rather than statically amphiphilic. In a sense, these metaphilic peptides are a molecular analog of recently engineered omniphilic/omniphobic surfaces [246–248]. Computer simulations indicate that the quasi-liquid surface of the peptide allows it to adapt to environmental change by rearranging the flexible side chains, a capability that plays a key role in enabling unusual interactions with membranes. Specifically, these metaphilic peptides are able to induce membrane-destabilizing curvature necessary for permeation, which we determine using X-ray measurements. Furthermore, because these metaphilic molecules can adapt their surface chemistry, we can control their charge and hydrophobicity over a broad range and still maintain water solubility, unlike many AMPs and CPPs [142, 249–252]. This allows us to

show how the activity of these metaphilic peptides is amplified with hydrophobicity and cationic charge, and we rationalize these results using a quantitative mean-field theory. One goal of this chapter is to develop a general conceptual vocabulary to analyze how molecules of different architectures beyond linear peptides interact with membranes, and how these architectures consequently allow small quantitative changes in structural parameters to lead to qualitative differences in membrane interactions.

4.2 Results and Discussion